Primary closure after carotid endarterectomy is not inferior to other closure techniques

Objective: Primary closure after carotid endarterectomy (CEA) has been much maligned as an inferior technique with worse outcomes than in patch closure. Our purpose was to compare perioperative and long-term results of different CEA closure techniques in a large institutional experience. Methods: A consecutive cohort of CEAs between January 1, 2000, and December 31, 2010, was retrospectively analyzed. Closure technique was used to divide patients into three groups: primary longitudinal arteriotomy closure (PRC), patch closure (PAC), and eversion closure (EVC). End points were perioperative events, long-term strokes, and restenosis >= 70%. Multivariate regression models were used to assess the effect of baseline predictors. Results: There were 1737 CEA cases (bilateral, 143; mean age, 71.4 +/- 9.3 years; 56.2% men; 35.3% symptomatic) performed during the study period with a mean clinical follow-up of 49.8 +/- 36.4 months (range, 0-155 months). More men had primary closure, but other demographic and baseline symptoms were similar between groups. Half the patients had PAC, with the rest evenly distributed between PRC and EVC. The rate of nerve injury was 2.7%, the rate of reintervention for hematoma was 1.5%, and the length of hospital stay was 2.4 +/- 3.0 days, with no significant differences among groups. The combined stroke and death rate was 2.5% overall and 3.9% and 1.7% in the symptomatic and asymptomatic cohort, respectively. Stroke and death rates were similar between groups: PRC, 11 (2.7%); PAC, 19 (2.2%); EVC, 13 (2.9%). Multivariate analysis showed baseline symptomatic disease (odds ratio, 2.4; P = .007) and heart failure (odds ratio, 3.1; P = .003) as predictors of perioperative stroke and death, but not the type of closure. Cox regression analysis demonstrated, among other risk factors, no statin use (hazard ratio, 2.1; P = .008) as a predictor of ipsilateral stroke and severe (glomerular filtration rate <30 mL/min/1.73 m(2)) renal insufficiency (hazard ratio, 2.6; P = .032) as the only predictor of restenosis >= 70%. Type of closure did not have any predictive value. Conclusions: In our study, baseline risk factors and statin use, but not the type of closure, affect perioperative and long-term outcomes after CEA.